ML Engineer, Large-Scale Pre-Training

Mindbeam; United States; On-site; USD 100,000 - 140,000; Full time

Job summary

Mindbeam is looking for an expert to design and optimize large-scale pre-training systems for their generative AI models. You will build scalable pipelines and implement distributed training strategies while collaborating closely with researchers.

Ideal candidates possess strong Python skills, experience with large-scale training, and familiarity with ML frameworks like PyTorch and TensorFlow. This role promises impactful innovations in AI development within a collaborative environment.

Qualifications

• 2+ years of experience with large-scale model training.
• Strong coding skills in Python.
• Experience with GPU scheduling and memory optimization.

Responsibilities

• Build scalable pre-training pipelines for foundation models.
• Implement distributed training strategies across GPUs/TPUs.
• Collaborate with researchers to develop production-ready workflows.

Skills

Large-scale model training

Distributed systems

Python

ML frameworks (PyTorch, TensorFlow, JAX)

GPU scheduling

Memory optimization

Docker

Kubernetes

Education

Bachelor’s, Master’s, or PhD in Computer Science or related field

Job description

Mindbeam is looking for an expert to design and optimize large-scale pre-training systems for their generative AI models. You will build scalable pipelines and implement distributed training strategies while collaborating closely with researchers.

Ideal candidates possess strong Python skills, experience with large-scale training, and familiarity with ML frameworks like PyTorch and TensorFlow. This role promises impactful innovations in AI development within a collaborative environment.